TARL: Transaction-Aware Reliable Ledgers for Executable Memory Management in Long-Term Agents

2026-08-04

AI Summary

Existing memory update mechanisms in long-horizon agents support only binary write/retain decisions, making them ill-suited for fine-grained operations such as addition, correction, rejection, or deferral, which often leads to memory contamination and error accumulation. This work proposes TARL, a novel framework that introduces, for the first time, a transaction-aware multi-state memory ledger. TARL models memory updates as five distinct executable operations and incorporates affected-memory identification, temporal scope parsing, and source reliability assessment to maintain separate states for accepted, pending, and rejected memories. Training via contrastive learning on memory state transitions, TARL substantially improves action prediction accuracy and memory recovery capability. Additionally, we introduce TARL-Mem, a new benchmark featuring fine-grained action labels and next-state objectives, which significantly enhances the robustness and reliability of memory systems in long-horizon agents.

Abstract

Persistent memory helps long-term agents retain knowledge, yet a single update error can repeatedly distort future retrieval and reasoning. Most existing systems reduce memory updating to a binary Write/Hold decision, which cannot distinguish whether new information should be added, ignored, used to revise an outdated belief, rejected as unreliable, or deferred for verification. These choices may share the same binary label while producing fundamentally different memory states. We introduce TARL, a memory state update framework that maps each statement to one of five executable actions. TARL identifies the affected memory, resolves its temporal scope, compares source reliability, and updates accepted, pending, and rejected ledgers. It is further trained by comparing the memory states produced by alternative update operations, encouraging the model to select the operation that leads to the correct result. We also introduce TARL-Mem, a benchmark with fine-grained action labels and next-state targets. Across in-domain, cross-source, temporal, counterfactual, and sequential evaluations, TARL improves action prediction and state recovery, reduces memory pollution, preserves conflicting evidence, and limits cumulative corruption. The complete model implementation is provided in the supplementary material.

Problem

• memory management
• long-term agents
• memory updating
• reliability
• executable actions

Innovation

• transaction-aware memory
• reliable ledgers
• executable memory management
• memory state update
• long-term agents